แก้ไข

แชร์ผ่าน


Microsoft Defender for IoT alerts

Microsoft Defender for IoT alerts enhance your network security and operations with real-time details about events logged in your network. Alerts are triggered when OT network sensors detect changes or suspicious activity in network traffic that needs your attention.

For example:

Screenshot of the Alerts page in the Azure portal.

Use the details shown on the Alerts page, or on an alert details page, to investigate and take action that remediates any risk to your network, either from related devices or the network process that triggered the alert.

Tip

Use alert remediation steps to help your SOC teams understand possible issues and resolutions. We recommend that you review recommended remediation steps before updating an alert status or taking action on the device or network.

Alert management options

Defender for IoT alerts are available in the Azure portal, OT network sensor consoles, and the on-premises management console. With Enterprise IoT security, alerts are also available for Enterprise IoT devices detected by Defender for Endpoint, in Microsoft 365 Defender.

While you can view alert details, investigate alert context, and triage and manage alert statuses from any of these locations, each location also offers extra alert actions. The following table describes the alerts supported for each location and the extra actions available from that location only:

Location Description Extra alert actions
Azure portal Alerts from all cloud-connected OT sensors - View related MITRE ATT&CK tactics and techniques
- Use out-of-the-box workbooks for visibility into high priority alerts
- View alerts from Microsoft Sentinel and run deeper investigations with Microsoft Sentinel playbooks and workbooks.
OT network sensor consoles Alerts generated by that OT sensor - View the alert's source and destination in the Device map
- View related events on the Event timeline
- Forward alerts directly to partner vendors
- Create alert comments
- Create custom alert rules
- Unlearn alerts
An on-premises management console Alerts generated by connected OT sensors - Forward alerts directly to partner vendors
- Create alert exclusion rules
Microsoft 365 Defender Alerts generated for Enterprise IoT devices detected by Microsoft Defender for Endpoint - Manage alerts data together with other Microsoft 365 Defender data, including advanced hunting

Tip

Any alerts generated from different sensors in the same zone within a 10-minute timeframe, with the same type, status, alert protocol, and associated devices, are listed as a single, unified alert.

  • The 10-minute timeframe is based on the alert's first detection time.
  • The single, unified alert lists all of the sensors that detected the alert.
  • Alerts are combined based on the alert protocol, and not the device protocol.

For more information, see:

Alert options also differ depending on your location and user role. For more information, see Azure user roles and permissions and On-premises users and roles.

Focused alerts in OT/IT environments

Organizations where sensors are deployed between OT and IT networks deal with many alerts, related to both OT and IT traffic. The amount of alerts, some of which are irrelevant, can cause alert fatigue and affect overall performance. To address these challenges, Defender for IoT's detection policy steers its different alert engines to focus on alerts with business impact and relevance to an OT network, and reduce low-value IT related alerts. For example, the Unauthorized internet connectivity alert is highly relevant in an OT network, but has relatively low value in an IT network.

To focus the alerts triggered in these environments, all alert engines, except for the Malware engine, trigger alerts only if they detect a related OT subnet or protocol. However, to maintain triggering of alerts that indicate critical scenarios:

  • The Malware engine triggers malware alerts regardless of whether the alerts are related to OT or IT devices.
  • The other engines include exceptions for critical scenarios. For example, the Operational engine triggers alerts related to sensor traffic, regardless of whether the alert is related to OT or IT traffic.

Managing OT alerts in a hybrid environment

Users working in hybrid environments might be managing OT alerts in Defender for IoT on the Azure portal, the OT sensor, and an on-premises management console.

Note

While the sensor console displays an alert's Last detection field in real-time, Defender for IoT in the Azure portal may take up to one hour to display the updated time. This explains a scenario where the last detection time in the sensor console isn't the same as the last detection time in the Azure portal.

Alert statuses are otherwise fully synchronized between the Azure portal and the OT sensor, and between the sensor and the on-premises management console. This means that regardless of where you manage the alert in Defender for IoT, the alert is updated in other locations as well.

Setting an alert status to Closed or Muted on a sensor or on-premises management console updates the alert status to Closed on the Azure portal. On the on-premises management console, the Closed alert status is called Acknowledged.

Tip

If you're working with Microsoft Sentinel, we recommend that you configure the integration to also synchronize alert status with Microsoft Sentinel, and then manage alert statuses together with the related Microsoft Sentinel incidents.

For more information, see Tutorial: Investigate and detect threats for IoT devices.

Enterprise IoT alerts and Microsoft Defender for Endpoint

If you're using Enterprise IoT security in Microsoft 365 Defender, alerts for Enterprise IoT devices detected by Microsoft Defender for Endpoint are available in Microsoft 365 Defender only. Many network-based detections from Microsoft Defender for Endpoint are pertinent to Enterprise IoT devices, such as alerts triggered by scans involving managed endpoints.

For more information, see Securing IoT devices in the enterprise and the Alerts queue in Microsoft 365 Defender.

Accelerating OT alert workflows

New alerts are automatically closed if no identical traffic is detected 90 days after the initial detection. If identical traffic is detected within those first 90 days, the 90-day count is reset.

In addition to the default behavior, you might want to help your SOC and OT management teams triage and remediate alerts faster. Sign into an OT sensor or an on-premises management console as an Admin user to use the following options:

  • Create custom alert rules. OT sensors only.

    Add custom alert rules to trigger alerts for specific activity on your network that's not covered by out-of-the-box functionality.

    For example, for an environment running MODBUS, you might add a rule to detect any written commands to a memory register on a specific IP address and ethernet destination.

    For more information, see Create custom alert rules on an OT sensor.

  • Create alert comments. OT sensors only.

    Create a set of alert comments that other OT sensor users can add to individual alerts, with details like custom mitigation steps, communications to other team members, or other insights or warnings about the event.

    Team members can reuse these custom comments as they triage and manage alert statuses. Alert comments are shown in a comments area on an alert details page. For example:

    Screenshot of the alert comments area.

    For more information, see Create alert comments on an OT sensor.

  • Create alert exclusion rules: On-premises management consoles only.

    If you're working with an on-premises management console, define alert exclusion rules to ignore events across multiple sensors that meet specific criteria. For example, you might create an alert exclusion rule to ignore all events that would trigger irrelevant alerts during a specific maintenance window.

    Alerts ignored by exclusion rules aren't shown on the Azure portal, sensor, or on-premises management console, or in the event logs.

    For more information, see Create alert exclusion rules on an on-premises management console.

  • Forward alert data to partner systems to partner SIEMs, syslog servers, specified email addresses and more.

    Supported from both OT sensors and on-premises management consoles. For more information, see Forward alert information.

Alert statuses and triaging options

Use the following alert statuses and triaging options to manage alerts across Defender for IoT.

When triaging an alert, consider that some alerts might reflect valid network changes, such as an authorized device attempting to access a new resource on another device.

While triaging options from the OT sensor and the on-premises management console are available for OT alerts only, options available on the Azure portal are available for both OT and Enterprise IoT alerts.

Use the following table to learn more about each alert status and triage option.

Status / triage action Available on Description
New - Azure portal

- OT network sensors

- On-premises management console
New alerts are alerts that haven't yet been triaged or investigated by the team. New traffic detected for the same devices doesn't generate a new alert, but is added to the existing alert.

On the on-premises management console, New alerts are called Unacknowledged.

Note: You might see multiple, New or Unacknowledged alerts with the same name. In such cases, each separate alert is triggered by separate traffic, on different sets of devices.
Active - Azure portal only Set an alert to Active to indicate that an investigation is underway, but that the alert can't yet be closed or otherwise triaged.

This status has no effect elsewhere in Defender for IoT.
Closed - Azure portal

- OT network sensors

- On-premises management console
Close an alert to indicate that it's fully investigated, and you want to be alerted again the next time the same traffic is detected.

Closing an alert adds it to the sensor event timeline.

On the on-premises management console, New alerts are called Acknowledged.
Learn - Azure portal

- OT network sensors

- On-premises management console

Unlearning an alert is available only on the OT sensor.
Learn an alert when you want to close it and add it as allowed traffic, so that you aren't alerted again the next time the same traffic is detected.

For example, when the sensor detects firmware version changes following standard maintenance procedures, or when a new, expected device is added to the network.

Learning an alert closes the alert and adds an item to the sensor event timeline. Detected traffic is included in data mining reports, but not when calculating other OT sensor reports.

Learning alerts is available for selected alerts only, mostly those triggered by Policy and Anomaly engine alerts.
Mute - OT network sensors

- On-premises management console

Unmuting an alert is available only on the OT sensor.
Mute an alert when you want to close it and not see again for the same traffic, but without adding the alert allowed traffic.

For example, when the Operational engine triggers an alert indicating that the PLC Mode was changed on a device. The new mode might indicate that the PLC isn't secure, but after investigation, it's determined that the new mode is acceptable.

Muting an alert closes it, but doesn't add an item to the sensor event timeline. Detected traffic is included in data mining reports, but not when calculating data for other sensor reports.

Muting an alert is available for selected alerts only, mostly those triggered by the Anomaly, Protocol Violation, or Operational engines.

Tip

If you know ahead of time which events are irrelevant for you, such as during a maintenance window, or if you don't want to track the event in the event timeline, create an alert exclusion rule on an on-premises management console instead.

For more information, see Create alert exclusion rules on an on-premises management console.

Triage OT alerts during learning mode

Learning mode refers to the initial period after an OT sensor is deployed, when your OT sensor learns your network's baseline activity, including the devices and protocols in your network, and the regular file transfers that occur between specific devices.

Use learning mode to perform an initial triage on the alerts in your network, learning those you want to mark as authorized, expected activity. Learned traffic doesn't generate new alerts the next time the same traffic is detected.

For more information, see Create a learned baseline of OT alerts.

Next steps

Review alert types and messages to help you understand and plan remediation actions and playbook integrations. For more information, see OT monitoring alert types and descriptions.